cpen {simest}

C code for convex penalized least squares regression.

Description

This function is only intended for an internal use.

Usage

cpen(dim, t_input, z_input, w_input, a0_input,
	lambda_input, Ky_input, L_input, U_input,
	fun_input, res_input, flag, tol_input, 
	zhat_input, iter, Deriv_input)

Arguments

dim

vector of sample size and maximum iteration.

t_input

x-vector in cvx.pen.reg.

z_input

y-vector in cvx.pen.reg.

w_input

w-vector in cvx.pen.reg.

a0_input

initial vector for iterative algorithm.

lambda_input

lambda-value in cvx.pen.reg.

Ky_input

Internal vector used for algorithm.

L_input

Internal vector. Set to 0.

U_input

Internal vector. Set to 0.

fun_input

Internal vector. Set to 0.

res_input

Internal vector. Set to 0.

flag

Logical for stop criterion.

tol_input

tolerance level used in cvx.pen.reg.

zhat_input

Internal vector. Set to zero. Stores the final output.

iter

Iteration number inside the algorithm.

Deriv_input

Internal vector. Set to zero. Stores the derivative vector.

Details

See the source for more details about the algorithm.

Value

Does not return anything. Changes the inputs according to the iterations.

Author(s)

Arun Kumar Kuchibhotla, [e-mail].

Source

Dontchev, A. L., Qi, H. and Qi, L. (2003). Quadratic Convergence of Newton's Method for Convex Interpolation and Smoothing. Constructive Approximation, 19(1):123-143.